The first experiment was performed less than one month after collection, and the last experiment was done six months after collection.For each experimental replicate within an experiment , 6 seeds from each field-collected tomato type were placed into sterile 1.5 mL Eppendorf tubes and submerged in 400µl of sterile 10 mM MgCl2 solution and sonicated for 15 mins in a Branson M5800 sonicator. This sonicating water bath is different from laboratory sonicators used to disrupt cells; instead, these baths will dislodge bacteria with minimal disruption of their cell integrity. The liquid was then transferred into new sterile tubes and used as seed microbiome inocula. Prior to inoculation, seeds were surface sterilized using the following procedure: Seeds were first soaked in 2.7% bleach solution for 20 minutes, then washed with sterile ddH2O three times to remove any excess bleach. The last washes were plated on KB agar plates and incubated at 28 °C for 24 hours. After sterilization, 40µl of the original seed wash was pipetted directly on top of each individual seed. We did this so that each seed would receive roughly the same number of microbes that was removed during the sonication step. The removal and re-addition process was done, in general, so that every seed used in the experiment would undergo the exact same procedure, and the only difference would be receiving microbiota or not. Negative control seeds were each inoculated with 40µl of 10 mM MgCl2.In order to culture bacteria from the seeds used in this experiment, seeds were sonicated into sterile buffer, as above. Next, cannabis storage the seed wash was diluted 1:10 in sterile 10 mM MgCl2 solution and plated onto KB agar and Lysogeny Broth agar.
They were incubated for 48 hours at 28 °C. We were only able to culture bacteria from tomato types 4 and 2. On average, we cultured 40 colony- forming units from each TT4 seed. To isolate individual strains from the microbial community, we picked morphologically distinct colonies, based on color and surface, and streaked them on new nutrient agar where they were grown for 24 hours at 28 °C. Liquid cultures were attained by inoculation into liquid KB and grown on an orbital shaker at 28 °C overnight.For consistency amongst tomato plant hosts, Money Maker seeds were used for all further experiments measuring the impact of particular seed-associated microbiota. Seeds were sterilized as described above. In addition to testing our own bacterial isolates ZM1, ZM2, and ZM3, we also included Biological Control strains, kindly provided by Dr. V. Stockwell, Oregon State. These two strains are Pantoea agglomerans strain E325A and Pantoea vagans strain C9-1. Bacterial inoculum was prepared as follows: isolates were grown overnight on an orbital shaker in LB at 28 °C. We measured the optical density at 600 nm of the overnight culture and plated the culture on LB agar, incubated overnight at 28 °C to obtain their CFU counts. The remainder of the liquid culture was stored at 4 °C overnight. The next day, we calculated a CFU to OD ratio, and re-measured the OD to account for any growth that occurred of the liquid culture overnight. We pelleted the bacteria at 4000 X G for 5 minutes, re-suspended in sterile 10 mM MgCl2 solution, and diluted to the appropriate concentration. Each seed was inoculated by pipetting the bacterial culture directly on top of each individual seed. In Figure 2-6, we inoculated seeds with pure cultures at a final inoculum density of 40 CFU/seed to approximately match the observed natural densities.
Each experimental replicate held four seedlings, and we had three experimental replicates per isolate per treatment. Disease severity was monitored for 10 days after plate flooding . For dose response curves, the density of bacteria applied to the seeds ranged from 4×10-1 to 4×106 CFU/seed with control replicates not receiving any, and disease was monitored for nine days. We did not replicate at the plate level for dose response curves.Pseudomonas syringae density was quantified from each experimental replicate using droplet digital PCR using a fluorescent probe targeting the Pseudomonas 16S gene as fully described elsewhere. Briefly, seedling homogenates were diluted 1:10, and 2µl of homogenate was used as template in the BioRad QX200 ddPCR reaction. In analyzing positive droplets, all thresholds were set using negative, no template controls and positive pure Pst DNA controls. As with analysis of AUDPC, Pst densities are a measure of each plate experimental replicate, as described above. Bacterial abundances are normalized to total seedling weight within a plate and reported as copy of 16S rRNA gene per gram of plant material. For negative ddPCR controls, we always attempted to measure Pst in the MgCl2 inoculated plant controls for all experiments as well as Pantoea DNA. Although the Pst probe was designed to be specific to Pseudomonads, we did this to ensure our probe was only amplifying Pst and not Pantoea nor any plant material. The signal amplitudes for sterile plant-only and Pantoea isolates-only controls were the same as those of no-template, sterile ddH20 controls, indicating that indeed, there was no detectable background amplification of Pantoea species when using the Pseudomonas probe, and no Pseudomonas was present in the negative controls.
Using pure DNA of individual bacterial isolates, we amplified and sequenced 16S genes using 27R and 1492R primers . In addition to 16S, we sought to further discriminate against our potential strains, and so we amplified the gyrB gene and rpoB genes with previously published primers and PCR protocols. We performed a BLASTn search of all isolate sequences and recorded the top hits with the highest identity . Phylogenetic tree of isolates and neighbors were built using gyrB sequences. We placed our isolates within a subset of samples previously mapped in a Pantoea phylogenetic tree by Rezzonico et al.. Dr. T. Smits kindly provided the E325 gyrB sequence. The evolutionary history of our isolates and other strains was inferred by using the Maximum Likelihood method based on the Tamura-Nei model. The tree with the highest log likelihood is shown. Initial trees for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood approach, and then selecting the topology with superior log likelihood value. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 13 nucleotide sequences. All positions containing gaps and missing data were eliminated. There were a total of 316 positions in the final dataset. Evolutionary analyses were conducted in MEGA7.MiSeq sequencing files were demultiplexed by QB3 sequencing facility. Reads were combined into contigs using VSearch, and the remainder of the analysis was carried out in Mothur version 1.41.3 following their MiSeq SOP. Data were quality-filtered by length, ambiguous bases, and homopolymer length using the recommended Mothur parameters. Chimeras were removed using UChime. We used a 99% similarity cut-off for defining OTUs. The Silva reference database was used for sequence alignment and taxonomic assignment. Archaeal, chloroplast, mitochondrial and unknown domain DNA sequences were removed. Once an OTU table was generated in Mothur, the remainder of the analysis was performed in R using the Phyloseq package version 1.19.1 and Vegan package version 2.4-5. To account for reagent contaminants, we also sequenced DNA extraction kit controls and PCR controls along with our samples. Contaminant OTUs from control samples that were at a similar or higher relative abundance in control samples compared to experimental samples were removed from the full OTU table. Data were rarified to 50,000 reads per sample and singletons were removed.The field of microbiome science spans both basic and applied research in human health, agriculture, and environmental change. As our understanding of the ability of the microbiome to influence host health and shape host traits deepens, curing bud there is increasing interest in selecting and/or designing microbiomes for specific traits or functions. Such trait-based selection of microbiomes has the potential to shape the future of agriculture and medicine. In agriculture, below ground microbiota have already proven capable of shifting the flowering time of plant hosts, enhancing drought resistance, and even altering above-ground herbivory. However, long-term, repeatable success of future efforts will rely on a fundamental understanding of the assembly of, selection within, and co-evolution among microbiota within these communities. One of the challenges facing successful, rational microbiome manipulation and assembly is disentangling the forces naturally shaping the communities, including both host characteristics and constant microbial immigration on community stability. For example, in both humans and plants, there is contrasting evidence for the relative importance of the environment versus host genotype in shaping the microbiome, and dispersal has been shown to override host genetics in an experimental zebra fish system. One powerful but under-utilized approach to understand and experimentally control for the factors shaping microbiome composition and diversity is experimental evolution.
Measuring changes of populations or communities over time under controlled settings in response to a known selection pressure has proved a powerful force in gaining fundamental understanding of both host-pathogen evolution and microbial evolution. Here, we harness an experimental evolution approach in order to study how an entire microbial community can be selected upon in a plant host environment that varies across disease resistance-associated genotypes. We employ a microbiome passaging approach using the phyllosphere microbiome of tomato as a model system to determine if the microbial community could become adapted to the plant host environment. The phyllosphere, defined as the aerial surfaces of the plant, is a globally important microbial habitat. Microbial communities in this habitat can shape important plant traits such as protection against foliar disease and growth. Successful traitbased selection on the phyllosphere could therefore allow for enhancement of plant health, but this critically depends on the ability to select for a well-adapted microbial community that is relatively stable against invasion. We collected a diverse phyllosphere microbiome from tomatoes grown in an agricultural setting and transplanted it onto green-house grown plants using a transplantation method previously shown to be effective for lettuce [118]. We serially passaged this diverse microbiome on each of four cohorts of tomato plants of five different genotypes for a total of 30 weeks. We then measured adaptation of the community both computationally by fitting community structure to neutral models, and empirically using community coalescence experiments in which communities from different passaged lines are combined together and re-inoculated onto host plants in a common garden experiment. Overall, we were able to measure and characterize the response of the phyllosphere microbiome to selection in the plant host environment under greenhouse conditions, and select for a stable and well-adapted plant-associated microbiome.A diverse starting inoculum was collected from field grown, mature tomato plants. This field-microbiome was spray inoculated onto 30 tomato plants of 5 different genotypes, with six replicates each. Two-week old tomato plants were spray-inoculated once per week for five weeks, and then sampled in their entirety ten days after the final inoculation . The phyllosphere microbiome of each plant was then individually passaged on these genetically distinct hosts over the course of four eight-week long passages; P1, P2, P3, and P4 . Microbiomes were not pooled across plants within a given plant genotype, resulting in 30 independent selection lines. Control plants were inoculated with an equal volume of either heat killed inoculum or sterile buffer every week. At the end of each passage, bacterial density was measured and normalized to the weight of each plant , and communities were sequenced using 16S rRNA amplicon sequencing. We first measured the impact of host genotype on bacterial community structure . Using Bray-Curtis dissimilarity measures, we performed an ANOSIM test and found that plant genotype could explain 29% of dissimilarity between microbiomes in P1 . In P2, plant genotype similarly explains 28% of the variation in bacterial community dissimilarity . However, genotype becomes an insignificant driver of community composition in both P3 and P4 . The genotype effect observed in P1 was robust to removal of the primary outlying line , and that same line had too low read depth to be analyzed at P2, and thus was excluded from this analysis at the rarefaction step. By P3, this line was included, as it did not fall outside of the 95% confidence intervals for P3 clustering. We also sought to determine if there were more subtle influences of host genotype on the community that were not uncovered through analyzing Bray-Curtis distances alone.